MP47-11: Differential Association of Body Mass Index on Survival Outcomes in Renal Cell Carcinoma: Impact of Tumor Histology

Arman Walia*, San Diego, CA, Yasuhisa Fujii, Tokyo, Japan, Viraj Master, Atlanta, GA, Ava Saidian, Rekha Narasimhan, Mimi Nguyen, John Perry, Margaret Meagher, Madison Chakoumakos, Kevin Hakimi, San Diego, CA, Dattatraya Patil, Atlanta, GA, Yosuke Yasuda, Hajime Tanaka, Tokyo, Japan, Ithaar Derweesh, San Diego, CA

Introduction: The “obesity paradox” for renal cell carcinoma (RCC) has been described in clear cell histology. We investigated impact of Body Mass Index (BMI) on outcomes in Clear Cell (ccRCC) and Non-Clear Cell RCC (nccRCC).

Methods: The International Marker Consortium for Renal Cancer database was retrospectively analyzed for patients with ccRCC and nccRCC (papillary and chromophobe). Adult BMI (kg/m2) cutoffs were defined as obese >30 and non-obese =30. Descriptive analyses were conducted for demographics and disease characteristics. Multivariate analysis (MVA) via Cox regression and Kaplan Meier analyses (KMA) were conducted for survival assessment. Primary outcome was risk of cancer-specific mortality (CSM); secondary outcome was all-cause mortality (ACM).

Results: 3,742 patients were included (ccRCC n=2,895/nccRCC n=847; median follow up 107 months). BMI >30 was independently associated with lower risk of CSM in ccRCC patients (HR 0.69, p=0.038). Hypertension (HR 2.29, p<0.001), high grade (HR 3.27, p<0.001), and stage 3/4 (HR 5.29, p<0.001) were associated with increased risk of CSM. For nccRCC, stage 3/4 disease (HR 19.48, p<0.001) was associated with worsened CSM, while BMI >30 was not a risk factor (HR 1.38, p=0.415).

BMI>30 was independently associated with decreased ACM in ccRCC patients (HR 0.74, p=0.011). Age (HR 1.03, p<0.001), male gender (HR 1.39, p<0.001), hypertension (HR 1.39, p<0.001), high grade (HR 1.72, p<0.001), and stage 3/4 (HR 2.33, p<0.001) were associated with increased risk of ACM, while increasing preoperative eGFR (HR 0.99, p<0.001) was protective. BMI >30 was not a risk factor for ACM in nccRCC (HR 0.99, p=0.99). Figure 1 provides KMA survival curves for cancer-specific (CSS) and overall survival (OS). For ccRCC stratified by BMI >30 and =30, 5-year CSS rates were 88% vs 88% (p=0.686) and OS 77% vs 73% (p=0.022), respectively. For nccRCC, 5-year CSS rates were 91% vs 93% (p=0.620) and OS 78% vs 80% (p=0.180) for BMI >30 and =30, respectively.

Conclusions: BMI >30 was independently associated with decreased risk of mortality outcomes in ccRCC but not nccRCC. Mechanisms are unclear but call for further investigation into impact of BMI and metabolic sequelae on different RCC histologic subtypes.

Source of Funding: None